printable banner

U.S. Department of State - Great Seal

U.S. Department of State

Diplomacy in Action

27. Letter of William H. Taft, IV, Legal Adviser of the Department of State to Senator Byron L. Dorgan of North Dakota regarding a Memorandum of Understanding signed by the State of Missouri and the Province of Manitoba, November 20, 2001


November 20, 2001

Dear Senator Dorgan:

This is in response to your February 14, 2001 letter requesting the Department of State’s views on the Memorandum of Understanding (MOU) signed by the State of Missouri and the Province of Manitoba. I apologize for the delay in responding. As described in the attached memorandum, the Constitution does not assign to the Executive Branch the responsibility for interpreting or enforcing the Compact Clause. However, the way this clause is applied has obvious effects on Executive Branch interests in the field of foreign affairs, and it is not uncommon for states of the United States to consult with the Department of State when they are considering entering into arrangements with a foreign power that may engage these interests. In the case of the MOU you have brought to our attention we were not consulted prior to its signature. The attached memorandum describes some of the considerations that we would have raised if we had been consulted.

I trust that you will find the attached Memorandum responsive to your inquiry.


William H. Taft, IV

As stated

The Honorable
Byron L. Dorgan,
United States Senate.


This memorandum sets forth Department of State comments on the January 25, 2001 Memorandum of Understanding between the State of Missouri and the Province of Manitoba (“MOU”) in light of relevant provisions of the U.S. Constitution.

In the MOU, Missouri and Manitoba agree “to work cooperatively to the fullest extent possible consistent with law and existing treaties . . . in their efforts to oppose water transfers” between the Missouri River watershed (Missouri’s water supply) and the Hudson Bay watershed (Manitoba’s water supply). The MOU includes commitments to exchange information; to mutually support opposition to inter-basin transfers, including related incremental works; and to communicate concerns about such transfers to their respective national governments.

There appear to be three constitutional doctrines implicated by the MOU: (a) the Compact Clause; (b) the Supremacy Clause by which federal law may preempt state action; and (c) the Foreign Affairs Power generally.

The MOU and the Compact Clause

The question has been raised whether the MOU, given that it has not been approved by Congress, is consistent with the Compact Clause of the Constitution. Article 1, Section 10, Clause 3 of the Constitution provides that "[n]o State shall, without the Consent of Congress ... enter into any Agreement or Compact with another State, or with a foreign Power ..." The Constitution does not specifically assign responsibility for interpretation or enforcement of this clause to the Executive branch of the federal government. In practice, however, it is not uncommon for states of the United States to consult with the Department of State when they are considering entering into an arrangement with a foreign power for advice as to the consistency of that arrangement with the Compact Clause. In the first instance, responsibility for fidelity to the requirements of the Compact Clause lies with the states themselves, pursuant to the Supremacy Clause of the Constitution. Should they submit a proposed compact to the Congress, it is the prerogative of the Congress to approve or disapprove the compact, or to require modifications. Ultimately, issues concerning the Compact Clause or a particular arrangement by a state with a foreign power may need to be resolved in the courts, either state or federal.

The Department of State has not been consulted by the state authorities of either North Dakota or Missouri concerning the MOU at issue here, and thus is not aware of whether there is an intention to bring the MOU before the Congress or the courts. However, in accordance with the Department’s normal practice, this memorandum identifies the kinds of considerations that the Department would raise about an MOU like this.

The Scope of the Compact Clause

The Department ordinarily looks to Virginia v. Tennessee, 148 U.S. 503 (1893), in assessing whether an agreement involving a U.S. state would constitute a “Compact ... with a foreign Power,” although that case did not involve a compact with a foreign power. The only Supreme Court case actually to review a potential state compact with a foreign power, Holmes v. Jennison, resulted in a divided court. The case involved the question of whether the Governor of Vermont had entered into an agreement with Canadian authorities to extradite a fugitive back to Canada. Chief Justice Taney, speaking for three other justices, took the view that “every agreement, written or verbal, formal or informal, positive or implied, by the mutual understanding of the parties” falls within the Compact Clause’s ambit. Taney was particularly concerned about the ability of a U.S. state to extradite fugitives to a country when it was the policy of the federal government not to extradite persons. In Taney’s view, the only permissible way for Vermont to make such a hand-over would be if Congress consented, since that would make the agreement subject to federal supervision. In contrast, the other justices found either that the Supreme Court had no jurisdiction to hear the case or that no agreement could be inferred.

In general, the notion articulated by Chief Justice Taney that all U.S. state agreements constitute compacts that require congressional consent has not been widely supported. In Virginia v. Tennessee, the Supreme Court, in reviewing an interstate compact delineating a boundary line, concluded that despite the Constitution’s general language, its prohibition on compacts without congressional consent was not absolute. Specifically, the Court reasoned the Clause should only extend to those compacts that involved “the formation of any combination tending to the increase of political power in the states, which may encroach upon or interfere with the just supremacy of the United States.” Subsequent Supreme Court case law has affirmed that, at least with respect to interstate compacts, only compacts that would increase the political power of the states in such a way as to interfere with the supremacy of the federal government require congressional consent.

Although it is not a settled question that the Virginia standard applies to state compacts with foreign powers, at least one state court, the Department of State and numerous scholars have assumed that it does. In McHenry County v. Brady, the Supreme Court of North Dakota declined to enjoin construction and maintenance of a drain from North Dakota into Canada called for by a contract between U.S. and Canadian municipal entities as a violation of the Compact Clause. In so ruling, the court declined to adopt the “sweeping language” of Jennison since the subject matter of that case (extradition) involved a national power, and instead relied on Virginia and its progeny in light of the local context in which the contract was concluded. In a similar 1981 case regarding a proposed international water district involving areas of both Vermont and Quebec, the Department of State took the view that such an arrangement did not implicate the Compact Clause because federal permitting procedures would still apply and the district’s activities would be limited to traditionally local functions (e.g., water service) rather than political functions.

In practice, Congress has been asked to consent to only a few foreign compacts involving U.S. states, leaving uncertain Congress’ view of the scope of the Compact Clause. However, we are aware of no compacts approved by the Congress that involved local interference with national policy. Among the most well-known examples of congressionally-approved compacts are a 1956 New York-Canada agreement to establish a port authority for a bridge across the Niagara river; a 1958 Minnesota-Manitoba highway agreement; 1949 and 1952 Northeastern Interstate Forest Fire Protection Compacts; and various compacts authorized under the 1972 International Bridge Act. In one case involving water rights, Congress consented in 1968 to a Great Lakes Basin Compact. Originally intended to include all U.S. states and Canadian provinces bordering the Great Lakes, the compact was to establish a Commission with the goal of promoting the use, development and conservation of the water resources of the Great Lakes. In giving its consent, however, Congress refused to approve certain compact provisions, including those that allowed Canadian provinces to join as members, in light of Department of State concerns about such participation and the potential overlap between the compact and the mechanisms established under the 1909 U.S.-Canadian Boundary Waters Treaty.

At the same time, the Department of State is aware that U.S. states often conclude various arrangements with foreign powers without congressional consent. It appears that such arrangements principally involve matters of common local interest, e.g., coordination on roads, police cooperation, border control, local trade cooperation initiatives, education exchanges, local conservation measures, and similar matters. When they are called to the Department’s attention, such arrangements have generally been analyzed under the Virginia standard, with particular attention to whether such texts would interfere with the President’s foreign relations responsibilities.

The MOU and the Compact Clause

Turning to the MOU, it appears that two questions need to be asked to determine whether it triggers the Compact Clause’s requirement for congressional approval. First, is the MOU a “compact or agreement” for constitutional purposes? Second, if so, does it belong to that class of agreements that the Supreme Court has determined require congressional consent?

As for the first question, to qualify as a “compact or agreement” the Department traditionally has looked to whether the text in question is intended to be legally binding. The form and the content of this MOU suggest that Missouri and Manitoba likely intended to conclude such a legal agreement. The MOU is structured as an agreement with a title, preamble, specific commitments and a signature block. The terminology used (e.g., “agree” and “ensure”) is consistent with a legally binding intent. A Missouri Department of Natural Resources Press Release calls the MOU an “historic agreement” that “commits both jurisdictions to working together to oppose water transfers between major watersheds.” Upon signing, Manitoba Premier Doer indicated that “today’s signing of this MOU commits both of our jurisdictions to work together to oppose any efforts that may result in the transfer of water between watersheds.” The two sides have also convened an inaugural meeting under the MOU to discuss their concerns over potential inter-basin water transfers.

The fact that the two parties condition their cooperation on existing law and treaties does not preclude a finding that the MOU is intended to be legally binding. The United States has concluded a number of treaties and other international agreements in which a particular provision or the agreement as a whole is subject to the parties’ laws or international commitments. In such circumstances, although the parties can avoid their obligations based on an existing law or treaty, they may not avoid such obligations simply because, from a policy perspective, they no longer desire to comply with them.

Ultimately, however, the legal status of an instrument such as the MOU may not itself be determinative of whether the document qualifies as a compact. As the Supreme Court reasoned in U.S. Steel Corp. “the mere form of the interstate agreement cannot be dispositive.” In other words, even in the absence of a legally binding agreement, the Compact Clause may be implicated. In Northeast Bancorp., Inc., for example, the Court undertook a Compact Clause analysis of reciprocal state banking legislation even where there was no evidence of a legal agreement between the states to enact such legislation. Instead, the Court looked for “several of the classic indicia” of a Compact: e.g., establishment of a joint organization or a body; some restriction on the state’s ability to withdraw from the arrangement by repealing or modifying its law unilaterally; or a requirement that limitations on state action are reciprocal. Although these factors seem particularly relevant where a court has to determine if independent state statutes constitute a compact, the Court has not to our knowledge addressed whether such indicia are also required where in fact a legal agreement exists. At a minimum, however, assuming that the same indicia applicable to interstate compacts apply to state compacts with foreign powers, these indicia are useful in evaluating the MOU.

Whether the “indicia” cited in Northeast Bancorp, Inc. are present in the MOU is not immediately apparent. Missouri and Manitoba have had at least one meeting "under" the MOU, but it is not clear if such meetings would constitute the “joint organization” referred to by the Supreme Court. Another question is whether Manitoba could argue that Missouri had violated the MOU if Missouri announced that it supported inter-basin water transfers (à la a repeal in legislation). Similarly, Northeast Bancorp, Inc. would ask whether the obligation of Missouri to cooperate in opposing inter-basin water-transfers is contingent on Manitoba's performance of similar obligations. Firm answers to such questions would require further factual development of what actions the parties understood as being required by their agreement “to work cooperatively to the fullest possible extent consistent with law and existing treaties . . . to oppose [inter-basin] water transfers.”

Assuming for purposes of analysis that the MOU constitutes a “compact or agreement,” the next question is whether it is the sort of compact or agreement for which congressional consent is required. As stated above, the Department traditionally applies the standard laid out in Virginia – i.e., whether a compact is “directed to the formation of any combination tending to the increase of political power in the states, which may encroach upon or interfere with the just supremacy of the United States.” Evidence of an actual impact on the federal government’s supremacy has traditionally not been required; it is the potential impact of the compact that has led the Department to point out the need for congressional consent.

Examining the MOU in light of Virginia and its progeny, the Department would look to whether the MOU (a) impacts other U.S. states; (b) interferes with the federal government’s interests in inter-basin water transfers; (c) deals solely with local matters; or (d) involves activities that could be carried out by Missouri even in the absence of the MOU. The following discussion briefly reviews each of these factors.

First, with respect to effects on other states, the water in the Missouri and Hudson Bay watersheds that is the subject of the MOU borders or supplies water for numerous states. Missouri and Manitoba are therefore not the only parties interested in how those watersheds are treated. Missouri’s alliance with Manitoba to support each other’s effort to oppose inter-basin water transfers could affect the interests of other states both as to the outcome and the process leading to decisions on how these waters are managed.

Second, in terms of the federal government’s role, Congress has indicated an express interest in the inter-basin transfers at issue in the MOU. Two statutes -- the Dakota Waters Resources Act of 2000 (“DWRA”), which amended the Garrison Diversion Reformulation Act of 1986 (“Garrison Act”) and the Garrison Act itself – address inter-basin water transfers directly. Pursuant to authorities in the Garrison Act, as amended, the Department of Interior, in consultation with the Department of State, recently approved construction of a relatively small-scale endeavor, the Northwest Area Water Supply (“NAWS”) project, which will result in transfers of water from the Missouri River watershed to the Hudson Bay watershed. In addition, the DWRA contemplates a potential future authorization of transfers between these watersheds on a larger scale. The DWRA provides a comprehensive set of procedures for the Secretary of the Interior to follow in order to study and possibly construct projects involving inter-basin transfers in the Red River Valley (part of the Hudson Bay watershed), with both federal and state involvement in the review process. Ultimately, the DWRA reserves to Congress the final decision on whether a transfer will be authorized, but any such transfers are limited to those that the Executive branch determines comport with the 1909 U.S.-Canada Boundary Waters Treaty’s restrictions on activities that might pollute or otherwise affect the level or flow of boundary waters.

Given such federal interest, application of the Virginia standard would require an analysis of whether the MOU encroaches on the political power of the federal government to address inter-basin water-transfers. It is not enough to show simply that the federal government has competence over these matters. Rather, one must ask whether Missouri’s enlistment of Manitoba’s support in the MOU to oppose particular transfers potentially operates to the legal detriment of the federal government by interfering with the decision-making scheme set out in the federal legislation or, where decisions have been made, in their effective implementation. A secondary question is whether the MOU could interfere with administration of the Boundary Waters Treaty. That Treaty affords the United States and Canada, not Missouri or Manitoba, the rights to interpret and apply the Treaty as well as to refer matters to the International Joint Commission.

As indicated above, a third factor the Department would customarily examine is the question of whether the MOU deals only with matters of local policy. Some state arrangements with foreign powers dealing with water use issues have been deemed to be solely of local interest for Compact Clause purposes. This was true of the drainage basin at issue in McHenry County and the Vermont-Quebec International Water District, which had “no political function.” The MOU in this case, in contrast, addresses cooperation between a U.S. state and a Canadian province to work together to oppose the possibility of inter-basin water transfers that could affect other states of the United States and which are to be considered pursuant to federal statute.

Fourth, the Department would assess the implications of the Supreme Court’s decision in Multistate Tax Commission, which highlighted that congressional consent to an interstate compact is not required so long as the state is free to undertake the contemplated activities in the absence of the MOU, on a proposed arrangement. Thus, if one could show in this case that activities contemplated under the MOU – i.e., sharing information on actions contemplated under the DWRA, opposing inter-basin water transfers and communicating concerns about such transfers to the federal government – are actions that Missouri has the authority to carry out irrespective of an MOU, it would argue against applying the Compact Clause.

A key inquiry for this purpose is the extent to which the MOU calls for “mutually supportive” cooperation which might be understood as cooperation that cannot occur without another party. This would pose two issues: first, the extent to which such activities are possible even in the absence of the MOU, and secondly, whether this kind of activity impinges upon the “exclusive foreign relations power expressly reserved to the federal government,” and therefore falls outside the Multistate Tax Commission authorization for interstate compacts to be concluded without Congressional approval.

Finally, in addition to these four factors, evidence of agreement on concrete actions by the parties undertaken pursuant to the MOU could assist in ascertaining whether the MOU impacts our federal structure. The MOU, however, is not so specific as to require either party to cooperate in ways that must physically manifest themselves (i.e., constructing a facility, etc.) nor does it appear to require them to enact any reciprocal obligations into law. This is presumably because the object and purpose of the MOU seems to be to commit Missouri and Manitoba to oppose the actions of others; i.e., to oppose what the federal government is studying, and in some cases, doing, with respect to inter-basin water transfers. Thus, any interference that the MOU might cause to the federal government’s supremacy would likely be procedural rather than substantive in nature. For example, if the MOU requires Missouri to operate not only on its own behalf, but also on Manitoba’s behalf, in attempting to influence federal water management policy, would that interfere with the federal government’s ability to implement the DWRA, the Garrison Act and the 1909 Boundary Waters Treaty? As discussed, below, in the event Missouri sought to undertake concrete actions with respect to such water management issues, a strong argument can be made that such actions would be pre-empted by the DWRA, the Garrison Act, and the 1909 Boundary Waters Treaty.

Because of the expressions of federal policy, in addition to Compact Clause considerations, the MOU also potentially implicates the more general constitutional issues of federal preemption and the foreign affairs powers of the federal government. This memorandum therefore provides some additional background on these separate constitutional issues.

The MOU, the Supremacy Clause and the Foreign Affairs Power

The Supreme Court decision in Crosby v. National Foreign Trade Council illustrates the Court’s most recent views on federal preemption in a foreign affairs context. In Crosby, the Court held unanimously that a Massachusetts law imposing sanctions on Burma was invalid under the Supremacy Clause of the Constitution “owing to its threat of frustrating federal statutory objectives.” In so holding, the Court concluded that a state law must yield to a congressional Act if Congress intends to occupy the field, even if the federal statute does not contain an express preemption provision. The Court did not base its holding on the federal government’s exclusive constitutional responsibility for foreign affairs, but it did reason that preemption was appropriate in part based on the state law’s disruption of the federal government’s ability to speak with one voice to foreign nations.

The Massachusetts law that was the subject of Crosby was enacted three months prior to a federal statute imposing mandatory and conditional sanctions on Burma. The Court found that the federal statute and the state law at issue had a common goal (economic sanctions against Burma based on its human rights record) and evidence was presented that it would not necessarily have been impossible for companies to comply with both the federal and state laws. But the Court found that the means employed by the commonwealth of Massachusetts were in conflict with those in the federal Act insofar as they were different and distinct from those prescribed by the federal statute, and that the common end did not neutralize the conflicting means. According to the Court, the fact that companies might have been able to comply with both sets of sanctions did not mean that the state Act was compatible with the federal Act, which gave maximum discretion to the President to calibrate the appropriate level of U.S. sanctions.

In examining the issue, the Court emphasized that “[i]t is simply implausible that Congress would have gone to such lengths to empower the President had it been willing to compromise the effectiveness by deference to every provision of state statute or local ordinance that might, if enforced, blunt the consequences of discretionary Presidential action.” Referring to the foreign affairs context of the statute, the Court also stressed that Massachusetts’ independent actions threatened the ability of the United States to speak effectively with one voice on the international plane, noting that “the President’s maximum power to persuade rests on his capacity to bargain for the benefits of access to the entire national economy without exception for enclaves fenced off willy-nilly by inconsistent political tactics.”

As far as the Department is aware the courts have not had occasion to consider the applicability of these principles to a state agreement with a foreign power, rather than a state statute. It would seem, however, that the logic of Crosby would prohibit states from accomplishing, via agreement with foreign states, what they are not able to accomplish by their own statutes. Therefore, it would appear relevant to assess the MOU’s operation in light of federal preemption principles.

The central issue would be the MOU’s compatibility with the federal statutory scheme for addressing the inter-basin water issues covered by the MOU. The NAWS project, for example, will involve such a transfer and has already been approved by the federal government in accordance with the terms of the Garrison Act. As for the DWRA, it is as comprehensive, if not more so, than the federal sanctions at issue in Crosby. Under the DWRA, the Secretary of Interior is charged with preparing a comprehensive report for Congress studying the Red River Valley’s water needs and options for fulfilling them. The Secretary is required to solicit input from states that may be impacted by possible options. Environmental impact assessments of all feasible options are mandated by the statute. Within this statutory scheme, the Secretary of Interior is given some responsibility for selecting among potential projects, with the notable exception that any project that would require transfer of water from the Missouri River or its tributaries must be submitted to Congress for specific approval by an Act of Congress.

Given this comprehensive scheme, it is plain that Congress intended, in enacting the DWRA, to ensure that the decision making process about water allocation to the Red River Valley be centrally coordinated at the federal level. State input is recognized by the DWRA as an important piece of the process, but it is clearly subsumed into a federal decision-making process that reserves all final decision-making authority to the federal government. Indeed it seems that one of Congress’ objectives was to reserve certain decisions not only to the federal government but to Congress’ power alone. In general, even where it does not reserve exclusive decision-making power for Congress, it is clear that the DWRA makes the issue of supply to the Red River Valley one of federal concern. The statutory scheme represents not merely a solution for a subset of issues related to the water needs of the Red River Valley, or a plan for addressing some specific geographic area representing part of the Red River Valley, but rather a complete plan for a federal approach to the total problem. As such, the statute appears to be designed to “occupy the field” when it comes to major decisions impacting certain water resources across several states.

Analogizing to the logic in Crosby, it is difficult to believe that Congress would have enacted the DWRA “had it been willing to compromise the effectiveness by deference to every provision of state statute or local ordinance” that might, if enforced, interfere with the overall purpose of the scheme. Any concrete actions by Missouri to oppose inter-basin water transfers outside of this scheme would likely be preempted in that they would interfere with federal policies and programs. On the other hand, Missouri is not precluded from expressing its own viewpoint on the resolution of federal water management issues; to the contrary, the DWRA explicitly allows Missouri such a role. Thus, the question under Crosby is whether through the MOU Missouri is seeking to afford a surrogate voice for Manitoba in the federal government’s decision-making and implementation processes that would interfere with the scheme envisioned by Congress.

Besides such principles of federal preemption, the courts have also confirmed the exclusive assignment of foreign affairs responsibilities to the federal government under the U.S. Constitution. Although the Court in Crosby did not reach the question of whether the Massachusetts statute unconstitutionally interfered in foreign affairs, both the district court and the appeals court held that it did, based on the decision by the Supreme Court in Zschernig v. Miller. The appellate court opined that “Zschernig stands for the principle that there is a threshold level of involvement in and impact on foreign affairs which the states may not exceed.” The court held that while the boundaries of Zschernig were unclear, the Massachusetts law was clearly inconsistent with the principle in Zschernig. The court rejected arguments by Massachusetts to the effect that courts must balance the interests in a unified foreign policy against the particular interests of an individual state. Rather, quoting from Zschernig, the court reiterated that “[state] regulations must give way if they impair the effective exercise of the nation’s foreign policy.” A similar ruling was recently issued by the U.S. District Court for the Northern District of California. In that case, the Court found Zschernig applicable where a state was conducting its own foreign policy.

Thus, depending on the extent of its actual interference with U.S. foreign policy efforts in managing the water resources of the Hudson Bay watershed shared with Canada, the Missouri-Manitoba MOU would need to be evaluated for whether it constitutes an unconstitutional disruption of the federal government’s foreign affairs power.


In light of the DWRA, the Garrison Act, the Boundary Waters Treaty and relevant practice, the Missouri-Manitoba MOU potentially implicates several constitutional doctrines. First, if the MOU is intended to be an instrument that could interfere with the just supremacy of the federal government, issues are raised as to the necessity for congressional consent under the Compact Clause. Given Congress’s occupation of the field of inter-basin water transfers by statute (e.g., the DWRA), there are further issues under Crosby which set out the standards for determining when a state statute is preempted under the Supremacy Clause. Finally, to the extent the MOU may potentially interfere with the foreign affairs power more generally it would need to be evaluated for its consistency with principles set out in Zschernig.

Back to Top

Do you already have an account on one of these sites? Click the logo to sign in and create your own customized State Department page. Want to learn more? Check out our FAQ!

OpenID is a service that allows you to sign in to many different websites using a single identity. Find out more about OpenID and how to get an OpenID-enabled account.